Videophone services have been growing popular. According to a mode of videophone services, animation characters and avatars are displayed on the videophone of a party who the speaker is talking to. An example of animation character that can be displayed on the videophone which is a mobile phone is shown in Non-patent Document 1.
Patent Document 1 also discloses a videophone for generating a virtual character based on the face of the speaker imaged by a camera and sending image data of the virtual character to a party who the speaker is talking to. According to the technology of Patent Document 1, the videophone recognizes which area represents a face from the face image captured by the camera, and extracts feature points including eyebrows, eyes, a nose, and a mouse from the face image to generate a virtual character.
An RGB format is known in the art for expressing image data. Known color spaces include an HSV color space, a uniform LCH color space, etc. Conversion between the RGB format and the HSV color space is disclosed in Non-patent Document 2, for example.
Non-patent Document 3 reveals conversion between the L*a*b* color model and the L*C*abhab color model, and conversion between the L*u*v* color model and the L*C*uvhuv color model. C*ab and C*uv correspond to the element “C” in the uniform LCH color space, and hab and huv to the element “H” in the uniform LCH color space.
The communication mode for displaying animation characters and avatars the videophone of a party who the speaker is talking to is problematic in that it is difficult for the party to identify the speaker.
Another problem is that the communication mode may adversely affect the communication depending on the party who the speaker is talking to. For example, if the party who the speaker is talking to is a speaker's superior such as a supervisor or the like, then the speaker's superior may judge the speaker as frivolous when an animation character or an avatar is displayed on the videophone of the speaker's superior.
In addition, the communication mode is also disadvantageous in that animation characters and avatars make it difficult to convey delicate facial expressions of the speaker and environmental information about the location of the speaker.
It is preferable to improve the above drawbacks for better communications.
According to the technology disclosed in Patent Document 1, the processing load is high because there is a need for a process for recognizing which area represents the face and extracting feature points. For conveying changes in the facial expression of the speaker in real time, a high-performance CPU is required and hence a cost increase incurred.    Patent Document 1: JP-A No. 2004-289254 (paragraphs 0017-0025)    Non-patent Document 1: “What 900i-series “Chara-den” can do”, [online], Jan. 23, 2004, IT Media Kabushiki Kaisha, [browsed Apr. 19, 2005], Internet <URL: http://www.itmedia.co.jp/mobile/articles/0401/23/news028.html>    Non-patent Document 2: Alvy Ray Smith, “COLOR GAMUT TRANSFORM PAIRS”, Computer Graphics, Vol. 12, pp. 12-19, 1978    Non-patent Document 3: Mikio Takagi, Akihisa Shimoda, “Image analysis handbook, new edition”, University of Tokyo Press, Sep. 10, 2004, p. 552-555